Omentum Preservation Versus Complete Omentectomy in Gastrectomy for Gastric Cancer

Primary objective:

The primary study objective is evaluate whether preservation of the omentum distal to the gastroepiploic vessels in gastrectomy for cancer is non-inferior to complete omentectomy in terms of three-year overall survival.

Secondary objectives:

Comparing the two study arms with regard to:

• Operating time
• Intraoperative blood loss
• Intraoperative complications
• Postoperative complications, defined according to the Clavien-Dindo classification25 and comprehensive complication index (CCI)
• Distribution of lymph node metastases
• R0-resection rate
• Rate of malignant cells in cytology
• Molecular sub classification of gastric cancer
• ICG fluorescent enhancement of omentum in omentum preservation group (in centers that have ICG fluorescence available)
• Protocol compliance to allocated treatment
• Hospital stay, defined as time interval between date of surgery and date of hospital discharge
• Readmission rate within 30-days after surgery
• Reintervention rate within 30-days after surgery
• Reoperation rate within three years after surgery
• Quality of life at baseline, 3, 6, 9, 12 and 24 months, the following questionnaires will be used: EQ-5D-5L, QLQ-C30, QLQ-OG25, CIPN, Happiness, HADS and work productivity
• 3- \& 5-year disease-free survival, defined as the period of time from operation to locoregional recurrence, peritoneal recurrence, distant metastases, second gastric cancer or death from any cause. Patients alive and free of all these events will be censored at the last follow-up.
• 5-year overall survival, defined as the period of time from operation to death from any cause. Patients alive and free of all these events will be censored at the last follow-up.
• Cost-effectiveness

Study design:

OMEGA is a randomized controlled, open, parallel, non-inferiority, multicenter trial. Eligible patients have to be operable (ASA \<4) with resectable (≦cT4aN3bM0) gastric cancer. Patients will be randomized in a 1:1 ratio between radical (sub)total gastrectomy with omentum preservation or complete omentectomy. Patients will be stratified according to center, neoadjuvant therapy and type of surgery (total or subtotal gastrectomy). The primary endpoint is overall survival at three-years after the operation. In total, 654 patients will be randomized.

Sample size:

The primary endpoint is three-year overall survival. According to survival numbers from the Dutch Cancer Registry (NKR), three-year overall survival after gastrectomy is approximately 50% in the Netherlands. Under the common assumption of exponential survival times, a hazard ratio of 0.862 under the alternative hypothesis, at least 50% and 45% expected events (i.e., death) in the control arm and experimental arm, respectively, at the minimum follow-up of three years, 298 events are needed in total to achieve 80% power at a one-sided significance level of 5% with a non-inferiority hazard ratio of 1.15 (PASS 15 Power Analysis and Sample Size Software (2017). NCSS, LLC. Kaysville, Utah, USA, ncss.com/software/pass), resulting in 314 patients per study arm. Dropouts will be rare (mostly due to loss to follow-up, which is quite rare in cancer patients), with proportion dropping out expected to be at most 5%. After correction for drop-out we plan to include 327 patients in each of the two arms (654 in total).

Statistical analysis:

Primary endpoint: Descriptive statistics will be calculated to summarize patients' groups included in each of trial arms.

Mean and standard deviation will be presented for normally distributed continuous variables. Median plus interquartile-range (IQR) will be presented continuous variables that are skewed and for ordinal variables. Dichotomous and nominal data will be summarized by means of frequencies and percentages.

Non-inferiority of the experimental treatment in terms of overall survival will be tested using Cox-regression. Non-inferiority will be concluded if the upper limit of the 90% confidence interval falls below the non-inferiority hazard ratio of 1.15, corresponding to a one-sided non-inferiority test at significance level of 5%. Survival will be presented graphically using Kaplan-Meier curves. All analyses will be according to the intention to treat principle. A per protocol analysis will also be performed. The experimental treatment will be declared non-inferior if non-inferiority is shown in both the intention to treat and the per protocol analysis.

Secondary endpoint: as independent samples t-test for normally-distributed continuous outcomes, Mann-Whitney tests for continuous outcomes that are not normally distributed or ordinal outcomes. Categorical outcomes will be compared using chi-square test or Fisher's exact test in case of low (expected) cell counts. Repeatedly measured outcomes will be compared between arms using linear mixed models. Secondary time-to-event outcomes will be compared the using log-rank test. Secondary endpoints will be tested at a two-sided significance level of 5%. Effect sizes suitable for the type of outcome measure will be provided (mean differences, ratio of geometric means, relative risks, hazard ratios) together with their 95% confidence interval.

Subgroup analysis for the effect of experimental treatment on overall survival will be performed for the follow subgroups: patient characteristics (age, male/female), diffuse/intestinal type gastric tumor, subtotal/total gastrectomy, and minimally invasive/open gastrectomy. Effect modification will use Cox regression with the subgroup variable, the arm and their two-way interaction. Additionally, stratified analyses will be performed where HR is calculated separately in each of the subgroups.

Quality of life data will be graphically represented across all time points and analyzed according to the manuals and will presented as domain and summarized scores. Questionnaire outcome comparisons will be analyzed using linear mixed models.